OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
11
An End to “Patience”?
The Great Recession and Economic
Protest in Eastern Europe
Mark R. Beissinger and Gwendolyn Sasse
This chapter explores the dynamics of economic protest across postcommunist Eastern Europe in the context of the Great Recession. According
to the IMF, as of 2010 the postcommunist countries were the world region
worst hit from the Great Recession, experiencing steeper economic declines
than any other part of the world (IMF 2010). Major economic contractions
occurred in seven postcommunist countries (Croatia, Estonia, Hungary,
Latvia, Lithuania, Slovenia, and Ukraine). By 2009 the GDP of Latvia was
22 percent lower than in 2007, while that of Lithuania had declined by 14 percent. From 2007 to 2009 unemployment rose in Hungary from 7.4 percent
to 10 percent, in Estonia from 4.7 percent to 13.7 percent, in Lithuania from
4.3 percent to 13.7 percent, and in Latvia from 6.0 percent to 17.1 percent. At
the same time, there was considerable variation within the region in the effects
of the recession. Four countries (Bulgaria, Czech Republic, Romania, and
Russia) experienced economic stagnation rather than negative growth, while
seven countries (Albania, Belarus, Macedonia, Moldova, Poland, Serbia, and
Slovakia) actually continued to achieve positive growth throughout the years
of the global recession.
Note: The authors thank Larry Bartels, Nancy Bermeo, and participants in the Oxford conference on Popular Reactions to the Economic Crisis for their comments on an earlier draft of this
chapter, as well as Tiffanee Brown, Lydia Dallett, Susan Divald, Andrea Hudecz, Seongcheol
Kim, Jana Pakstaitis, Olga Radchenko, and Bryn Rosenfeld for research assistance connected
with this project.
oxfordhb-9780199357505.indd 334
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
335
The Great Recession was not the first time in recent memory that postcommunist Europe went through a period of dramatic economic contraction.
The 1990s witnessed much more severe economic downturns for most countries of the region, as these societies underwent the transition from socialist to
capitalist economic systems. In the three years after 1991, GDP fell by 23 percent in Estonia and 18 percent in Latvia. Yet these two countries weathered
the transition to capitalism with less economic pain than many countries in
the region; the corresponding declines in GDP were much steeper in Russia
(37 percent), Ukraine (55 percent), and Moldova (59 percent). However, as
the literature on postcommunist economic transitions emphasizes, for the
most part the transition to capitalism, though wrenching, did not produce
major waves of protest in many countries, as citizens displayed a surprising
degree of “patience” (or at the very least, quiescence) in the midst of massive
economic decline. There were, of course, exceptions. Major waves of strikes
and labor protest occurred periodically in Russia in the 1990s, particularly
over the issue of unpaid wages (Robertson 2010). Poland has routinely been
singled out as an exceptional case of labor mobilization under socialism and
continued mobilization in the early postcommunist period (Ekiert and Kubik
1998, 1999; Seleny 1999). But in general, and contrary to what many analysts
had predicted, “patience” or quiescence is thought to have predominated over
protest, particularly in comparison with other regions of the world that also
experienced IMF-led structural adjustment programs (Greskovits 1998. See
also Przeworski 1991; Agh 1991; Haggard and Kaufman 1992; Walton and
Seddon 1994; Howard 2003; Ost 2005; Vanhuysse 2006).
The reaction to the Great Recession in Eastern Europe, however, indicates
that, at least in some parts of the region, this period of quiescence in response
to economic decline has come to an end. In 1992—a year in which the GDP
of Latvia contracted by 32 percent—only three known major demonstrations
involving thirty-six hundred participants (and no mass violence) occurred in
Latvia over economic conditions.1 By contrast, in 2009 Latvia experienced
thirteen major demonstrations over economic conditions, involving more
than fifty thousand participants and including significant violence resulting
in forty-three injuries, major property damage, and more than 250 arrests.
Indeed, as we will see, the Latvian example is no aberration; in a number of
the postcommunist countries the protests touched off by the Great Recession
were considerably greater than those that occurred in response to the economic downturn of the early 1990s.
In this chapter we explore the factors that shaped patterns of economic
protest across eighteen European postcommunist countries during the
oxfordhb-9780199357505.indd 335
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
336
M ass Polit ics in Tough T imes
Great Recession. We seek to explain how the Great Recession altered the
level and nature of economic protest in the region, and to explore the factors that explain the considerable variation in patterns of economic protest
across postcommunist Europe during this period. As we will show, contrary
to what one might expect, protest overall in the region declined during the
Great Recession. But this is not true of all forms of protest, or of protests
over all types of issues. Strikes, protests over ethnic and nationalist issues, and
protests for greater economic benefits declined sharply during this period.
By contrast, demonstrations assumed greater weight in protest repertoires,
and protests against economic cutbacks rose sharply to become the dominant
type of protest within the region. In this respect, the Great Recession altered
not only protest repertoires but also the character of economic protest, transforming it from a proactive endeavor demanding salary and benefit increases
to a more defensive one voicing discontent over austerity measures.
One might logically expect that those countries experiencing more economic pain would also be those experiencing greater economic protest, and
indeed we show that countries that continued to grow exhibited lower levels
of protest. But we also show that among those countries undergoing significant economic contractions, there was considerable variation in the extent
to which economic pain translated into economic protest. Ironically, those
countries that were most vulnerable to a high level of economic protest in the
late 2000s were those that had been in the forefront of economic reform in
the 1990s and most eager to integrate with Europe. This was so, we argue, for
two reasons: (1) they were more vulnerable than other countries of the region
to a serious economic downturn in the context of global recession, as a result
of their high dependence on the global economy; and (2) integration into the
EU and the factors that underlay it (as well as relatively successful patterns of
economic growth associated with it in the late 1990s) generated expectations
about an improving standard of living that were dashed in the context of the
Great Recession.
We also show that a number of other factors shaped the extent and nature
of protests in those states that experienced economic contractions: (1) level
of public sector employment, (2) IMF rescue packages, (3) public trust in
government in the run-up to the crisis, and (4) political party mobilization.
We illustrate these causal processes through paired comparisons of Latvia
and Estonia on the one hand and Hungary and Ukraine on the other. All
four countries experienced severe economic downturns in the context of the
Great Recession (among the seven most severe economic contractions in our
sample). But even though Latvia and Hungary were among the countries
oxfordhb-9780199357505.indd 336
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
337
exhibiting the highest levels of economic protest (in terms of number of protests, participation rate, and extent of violence) during this period, Estonia
and Ukraine remained relatively quiescent. We argue that these variations
in outcomes resulted from the countries’ differential reform trajectories and
how they interacted with varying degrees of trust in government and varying
abilities on the part of political actors to instrumentalize and channel political polarization.
Patterns of Economic Protest in Postcommunist
Countries
To analyze patterns of protest across European postcommunist states during
the global financial crisis, we constructed an event database of major protests and incidents of mass violence throughout the region time. Electronic
searches were conducted of five leading international newswires (Reuters,
Associated Press, Agence France Presse, Deutsche Presse Agentur, and
Interfax) for the period from January 2007 through December 2010. All articles that reported on demonstrations, strikes, or mass violent events in any
of the eighteen European postcommunist countries examined in this study
were saved in monthly media files for subsequent review and coding. (The
Appendix provides a more detailed description of our coding procedures.)
The utility of using five wire services is clear from the fact that no single wire
service covered more than 43 percent of the 967 protest events recorded in
this study. Reuters, which is often used in cross-national studies of protest
events (as, for instance, in the World Handbook of Political Indicators), covered only 17 percent of the protest events in the data. Our use of multiple
sources also increased the quality of the coded information for many events,
filling in details about individual events that otherwise would have remained
unknown had only a single source been available.
Of course, media studies of protest events are known to have significant
biases, and those biases are likely to be even more significant in a study based
on international wire services. Large numbers of protest events are ignored
in international wire service reports (especially small events and events outside capital cities), and any media-based study of protest activity is unlikely to
be a fully accurate reproduction of patterns of protest. It may, however, be a
representative sample of a certain sort. On the basis of what is known about
how the media report on protest events, one would expect that international
wire services would tend to report primarily on politically salient events and
oxfordhb-9780199357505.indd 337
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
33 8
M ass Polit ics in Tough T imes
on events involving violence or taking place in a country’s capital. These may
actually be the most important events for understanding the impact of economic decline on patterns of protest in any case, since they are the events
most likely to affect the political process. For purposes of obtaining a reasonably representative sample across eighteen countries, international wire services are probably the best sources available, since country-level sources have
quite different policies on reporting about protest events. It is possible—even
likely—that certain international wire services covered particular countries
more thoroughly than others. The use of reports from five international wire
services can serve to mitigate this selection bias. The sample is right-censored
in that both the Great Recession and economic protest in the postcommunist countries continued beyond December 2010, when our data collection
ended. But we began sampling in January 2007, well before the onset of the
recession in late 2008, in order to be able to identify how the recession itself
may have altered patterns of protest.
There is no scholarly consensus over how best to measure trends in protest action. A simple event count can be thought of as representing the frequency of attempts by movements to organize collective action. Often, this
is weighted by the number of days over which a protest event occurs, since
a one-day protest represents a qualitatively different protest effort from a
prolonged, multiweek action.2 But we are also interested in the resonance
of attempts to mobilize—in particular, the number of people who actually
participate in a protest event.3 We assume that the factors shaping movement attempts to mount protests are likely to differ from those that shape the
decisions of large numbers of people whether or not to participate in these
actions, given the opportunity to do so (Beissinger 2002).
We report all three measures (events, protest days, and protest participation) in Figure 11.1, which provides an overview of the evolution of protest
events in the region over time by form of action, as captured by the five international wire services. As is evident from the figure, during the period of the
Great Recession (and contrary to what one might expect) there actually was
a general decline in protest activity in the postcommunist region—irrespective of whether one measures protest by the number of events, protest days,
or participants. The decline began prior to the onset of the Great Recession
but continued to deepen in 2009 in the midst of the global financial crisis.
In terms of protest repertoires, most acts of protest reported by the wire services during this period (90 percent) involved demonstrations; only 11 percent involved strikes and only 9 percent involved mass violent events. (Protest
events in the sample could include more than one action form.) But even
oxfordhb-9780199357505.indd 338
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
1a: Number of protests
339
1b: Number of protest-days
350
700
300
600
250
500
200
400
150
300
100
200
50
100
0
0
2007
2008
2009
2010
2007
2008
2009
2010
1c: Participation in protests
5,000,000
4,500,000
4,000,000
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
0
2007
All protests
2008
2009
Demonstrations
2010
Strikes
Figure 11.1 Protest Activity by Form and Dimension of Mobilization, 2007–2010.
though they constituted a relatively small proportion of protest events overall, strikes accounted for 65 percent of all participants in protest actions during this period.
Over the course of the global financial crisis, the incidence of strikes (both
in terms of raw numbers and as a relative proportion of protest events each
year) declined sharply, as strikes became a diminishing part (and demonstrations became a larger part) of protest repertoires. Such patterns are in accord
with the findings of the scholarly literature on strikes, which has generally
found that strikes tend to be more frequent when economies are expanding
rather than contracting, since in tough times workers’ jobs are at the mercy of
employers, and workers have greater difficulty withholding labor when their
jobs are under threat. (See, for instance, Tilly and Shorter 1974; Hibbs 1976.)
As Figure 11.2 shows, much of the overall decline in protest activity in the
region during the Great Recession was driven by a decline in protests over noneconomic issues (in particular, ethnic and nationalist issues). This decline does
not appear to have been connected with the recession. In both Russia and
Hungary, for instance, 2007 (the year before the onset of the recession) saw
unusually intense ethnic and nationalist protest mobilization—in Russia, due
primarily to the conflict with Estonia over the removal of a Soviet war memorial
oxfordhb-9780199357505.indd 339
10/29/2013 11:10:17 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
340
M ass Polit ics in Tough T imes
2a: Number of protests
140
120
100
80
60
40
20
0
2007
2008
2009
2b: Number of protest-days
2010
450
400
350
300
250
200
150
100
50
0
2007
2008
2009
2010
2c: Participation in protests
4,000,000
3,500,000
3,000,000
2,500,000
2,000,000
1,500,000
1,000,000
500,000
0
2007
2008
Ethnic or nationalist issues
For new economic benefits
2009
2010
Economic issues
Against economic cutbacks
Figure 11.2 Protest Activity by Issue Area, 2007–2010.
and activity by extremist right-wing groups; and in Hungary, the result of the
activation of Viktor Orbán’s Fidesz party and protests by right-wing extremists against the socialist government of Prime Minister Ferenc Gyurcsány. The
latter were linked to the domestic economic crisis experienced by Hungary
prior to the onset of the global financial crisis and prepared the ground for
ongoing mobilization once external economic constraints grew.
Economic protest actually remained at roughly the same level during the
Great Recession as prior to the Great Recession, increasing as a proportion
of all protests during this period. Yet, the duration of economic protests and
the overall number of people who participated in them declined sharply during this period. Even so, economic issues throughout the Great Recession
resonated among protestors in ways that protest over other issues simply
did not. Thus, although constituting 47 percent of the number of protests
and 55 percent of the number of protest days, economic protest accounted
for 80 percent of the participants in protest during this period. Particularly
in 2009, immediately after the onset of the global financial crisis, the region
experienced an increase in the number of economically related acts of protest,
oxfordhb-9780199357505.indd 340
10/29/2013 11:10:18 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
34 1
amounting to 61 percent of all protests and 89 percent of all participants in
protests during that year.
Moreover, the Great Recession was associated with a sharp shift in the
nature of the economic issues raised in protests. In 2007, before the onset
of the economic crisis, three-fourths of all participants in protest acts in the
region mobilized in support of obtaining new economic benefits (increased
pay, improved working conditions, or improvements in services). By contrast,
in the wake of the global financial crisis, protest aimed at obtaining new benefits fell precipitously (declining to only 14 percent of protest participants in
2009 and 2 percent in 2010), while protest against economic austerity measures (cuts in pay, benefits, or services) came to dominate protest agendas,
with protests revolving around these demands mobilizing 71 percent of all
protest participants in the region in 2009 and 80 percent in 2010. In short,
the global recession fundamentally altered the character of protest politics
in the region, transforming protest into a largely defensive set of actions in
reaction to austerity measures that cut pay, benefits, and services rather than
revolving around gaining new benefits, as had overwhelmingly been the case
during more prosperous times.
These overall patterns hide considerable country-level variations, and it
is at the country level that one can identify some of the factors differentially
shaping protest responses across the region. Table 11.1 provides a general
overview of patterns of protest across the eighteen European postcommunist
states examined in this study, weighted by their population size.4 As can be
seen, not all countries experienced a high rate of protest over economic issues
during this period, and in some countries (Belarus and Estonia, for example)
protest over other issues predominated in the agendas. In this study we focus
on patterns of protest over economic issues rather than patterns of protest
in general, since we are interested specifically in probing the differential protest responses to the Great Recession, and as we have seen, economic protest
played the dominant role in the protest repertoires of the region during the
Great Recession. It is of course possible that there is an economic component
to protest over other types of issues (for instance, democratization or ethnic conflict) that we are bracketing by looking only at protest over economic
issues. However, we expect that protest over these other issues is likely to be
less directly connected with patterns of economic performance and more
likely to be driven by other factors, such as ethnic stratification and diversity and government repression, that are less relevant for understanding the
impact of the recession on patterns of protest.
oxfordhb-9780199357505.indd 341
10/29/2013 11:10:18 PM
oxfordhb-9780199357505.indd 342
1
52
16
7
2
1
47
50
94
14
2
9
68
454
9.4
9.2
15.0
2.8
5.8
1.4
1.9
3.2
1.8
5.9
1.1
4.5
3.8
0.8
5.2
7.3
2.0
1.0
0.2
1.2
2.3
0.7
1.9
0.4
4.5
1.5
1.0
6.8
3.1
1.7
0.7
1
37
13
4
0
0
19
31
43
2
1
4
18
206
5
13
7
8
0
0.8
3.7
5.9
1.1
0.0
0.0
0.5
1.4
0.3
0.3
0.2
2.0
0.4
0.5
1.8
1.6
0.8
0.0
7100
651453
74875
29500
349650
49183
1210550
1510955
1143226
292843
15948
899500
654255
10604421
46931
2308658
204010
1101484
54300
Protest
Participation
6
65
34
8
168
11
31
69
8
40
3
450
14
5
325
45
108
18
Protest
Participation per
1,000 Pop.
3000
557372
56175
25050
310000
100
1184850
1399905
253155
50000
5148
899500
224580
8521709
11185
2262287
178660
1077742
23000
2
56
25
7
149
0
31
64
2
7
1
450
5
1
319
40
106
8
3000
402622
42075
14650
0
0
560800
1138400
48766
10300
4500
572100
86017
3765305
5810
23098
37100
816067
0
2
40
19
4
0
0
15
52
0
1
1
286
2
1
3
8
80
0
0
3.091043
4.817617
3.931826
0
0
2.197225
1.791759
2.74084
2.944439
0
1.098612
1.386294
0.693147
4.034241
0
0
0
ParticiParticiParticiPartici- Economic
pation in
pation in
pation in
pation in Violence
Economic Economic
Protests
Protests
indexa
Protests Protests per Against
Against
1,000 Pop. Economic Cutbacks per
Cutbacks 1.000 Pop.
Note: a Calculated as the natural log of the product of the number of people injured in mass violent events concerning economic issues; * the score on a property
damage index for these same events.
10
48
14
17
3.1
9.7
6.5
4.5
Belarus
30
Bulgaria
69
Croatia
29
Czech
46
Rep.
Estonia
12
Hungary
92
Latvia
33
Lithuania 10
Macedonia 12
Moldova
6
Poland
72
Romania 70
Russia
247
Serbia
43
Slovakia
6
Slovenia
9
Ukraine
172
TOTAL 967
2
3.0
Total Protests Protests Economic Protests Protests
Protests per
Raising Protests Against Against
Million Economic per Economic Cutbacks
Pop. Demands Million Cutbacks per
Pop.
Million
Pop.
9
Albania
Country
Table 11.1 Country-level Indicators of Patterns of Protest, 2007–2010.
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
10/29/2013 11:10:19 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
3b: Growth and participation in economic protests
8
LV
BG
6
HU
SI
4
HR
2
LT
UA
EE
0
80
SI
6
Ln(Participation in econ
protests per 1000 pop)
Economic protests per million pop
3a: Growth and the number of economic protests
34 3
90
RO
CZ RS
MK PL BY
RU
AL
SK MD
100
110
BG
CZ
HU RO
HR
4
MK
PL
LV
LT
2
AL
RS
UA
EE
RU
SK
BY
MD
0
120
80
90
100
110
120
Economic growth, 2007-2010 (100=2007)
Economic growth, 2007-2010 (100=2007)
3c: Growth and economic violence
Economic violence index
5
LV
4
BG
LT
HU
3
RS
RU
PL
2
RO
UA
1
0
BY
HRCZ SKMKMD AL
EE
80
SI
90
100
110
120
Economic growth, 2007-2010 (100=2007)
Figure 11.3 Economic Growth and Economic Protest, 2007–2010.
In Figure 11.3 we use the population-weighted figures on protest activity to
examine the bivariate relationship between rates of economic growth and patterns of protest over economic issues, aggregated over the entire 2007–2010
period. In Figure 11.3a we focus on the number of events, and in Figure 11.3b
on the number of participants. In Figure 11.3c we focus on variation in the
level of mass violence over economic issues (as measured by an index reflecting the number of people injured in mass violent events and the level of
property damage involved in them).5 In all three figures, there is a negative
relationship between economic growth and level of economic protest, and
the patterns of variation in all three bear certain similarities. In Figure 11.3a,
for instance, those countries that experienced a positive rate of economic
growth during the Great Recession (Albania, Belarus, Macedonia, Moldova,
Poland, Slovakia, Serbia, and Russia) also had a lower number of economic
protests per population. But among countries that experienced zero or negative growth (Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia,
Lithuania, Romania, Slovenia, and Ukraine), there was a great deal of heterogeneity in the extent to which they experienced protest over economic
issues. Figure 11.3b shows a similar pattern for levels of participation, with
oxfordhb-9780199357505.indd 343
10/29/2013 11:10:19 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
34 4
M ass Polit ics in Tough T imes
a downward trend among those countries that experienced a higher rate of
growth, but with significant variation among countries that experienced zero
or negative growth. Latvia, Hungary, Bulgaria, Croatia, Slovenia, Lithuania,
and Romania exhibited the highest number of economic protests, while
Bulgaria, Slovenia, Macedonia, Czech Republic, Romania, Hungary, Poland,
and Latvia displayed the highest levels of participation in economic protests.
In Figure 11.3c, the overall pattern is again sharply negative, with greater variance in level of violence among countries that experienced zero or negative
economic growth. Latvia, Bulgaria, Lithuania, and Hungary stood out for
their relatively high levels of economically related violence during this period.
Thus we are left with a puzzle: economic growth may explain why those
societies that continued to benefit from growth during the Great Recession
did not experience a high level of economic protest, but it cannot explain
why, among those societies that experienced significant economic downturns
during the Great Recession, some exhibited considerable economic protest
while others did not. Countries experiencing similar amounts of economic
pain exhibited significantly different degrees of protest over the consequences
of economic contraction. We now turn to a multivariate analysis of patterns
of economic protest and to two paired case studies to help explain this puzzle.
Multivariate Analysis of Patterns of Protest
In the analysis that follows, we use country-years as the basic unit of analysis
and implement a cross-sectional time-series design, focusing on four dependent variables of interest: the number of protests (weighted by their duration)
over economic issues, the number of participants in protests over economic
issues, the number of protests (weighted by duration) specifically against
economic cutbacks, and the number of participants in protests specifically
against economic cutbacks.6 As we saw earlier, economic protest during
the 2007–2010 period varied considerably in its goals. In the context of the
Great Recession, protest for new economic benefits receded as the predominant form of economic protest, with protest over economic cutbacks instead
coming to dominate agendas. By exploring the factors shaping protest over
economic issues more broadly and against economic cutbacks more specifically, we can gain a good sense of some of the key drivers during the Great
Recession.
For the event-count dependent variables, we use a random effects negative
binomial model.7 For the participation dependent variables we use a tobit
model, which addresses the issue of nonrandom selection when a variable is
oxfordhb-9780199357505.indd 344
10/29/2013 11:10:19 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
34 5
censored or truncated at a certain point. In this case, one cannot have participation in protest without first having a protest, so that a simple regression
model confounds the issue of how many people decide to participate in protest with the issue of whether or not a protest is organized in the first place.
Participants in protest generally do not decide whether and when an event
will occur, but only whether to participate in an event already organized by
others. By censoring those observations in which no protests occurred, we can
estimate whether a particular independent variable increased or decreased
participation, given the availability of an opportunity for people to participate in protest.8
We begin with baseline models relating population size and rates of economic growth to the number of economic protests (in Table 11.2) and the
number of participants in those protests (in Table 11.3). As reported in Model
1 in Table 11.2a, both variables have statistically significant relationships with
the number of protests over economic issues in general. As one would expect,
countries with larger populations experienced more economic protest, and
economic growth was strongly related to lower levels of economic protest.
Controlling for population size, we found that every percentage point of
increase in economic growth (relative to the base year of 2006) decreased the
expected number of economic protests annually by 5.5 percent. As Model 1 in
Table 11.2b indicates, population size and rate of economic growth were also
strongly related to protest against economic cutbacks specifically. Every percentage point of economic growth (relative to the base year of 2006) decreased
the expected number of protests over economic cutbacks annually by almost
8 percent. Thus, in a country that experienced an economic decline of 14 percent below the 2006 level (as occurred, for instance, in Latvia in 2009), one
would expect a 77 percent increase in the rate of economic protest in general and a 109 percent increase in the rate of protest over economic cutbacks,
whereas in a country that experienced 14 percent economic growth compared
to the 2006 level (as occurred, for example, in Poland in 2009), one would
expect equivalent decreases in economic protest and protests against economic cutbacks.
Model 1 in Table 11.3a shows that population size had a marginally significant relationship with the number of people who participated, given the
opportunity to participate in a protest over economic issues, but economic
growth did not. However, as Model 1 in Table 11.3b indicates, given the opportunity to participate in a demonstration against economic cutbacks, both
population size and economic growth had statistically significant effects on
the number of people who participated. Every percentage point of economic
oxfordhb-9780199357505.indd 345
10/29/2013 11:10:19 PM
oxfordhb-9780199357505.indd 346
10/29/2013 11:10:20 PM
IRR
z−score
Model 1
IRR
z−score
Model 2
IRR
z−score
Model 3
IRR
5.97****
−3.62****
−1.64
−1.37
z−score
Model 4
Ln population
2.051
6.17**** 2.136
6.25**** 2.046
5.96**** 2.260
GDP growth (2006 = 100)
0.945
−3.83**** 0.949 −3.46**** 0.943
−3.57**** 0.943
Freedom House score, 2006
0.952 −0.94
0.878
EBRD economic reform index,
0.847 −0.49
0.528
1998
Public sector employment
Dummy, IMF standby program
Number of observations
72
72
68
68
Number of groups
18
18
17
17
Wald chi-square
42.17****
44.01****
40.14****
40.82****
Log likelihood
−211.612
−211.180
−200.880
−199.421
Variable
All Economic Protest
Table 11.2a: All Economic Protest Events
−0.53
0.825
1.002
0.09
0.897 −0.36
68
17
40.14****
−200.811
5.97****
−3.47****
z−score
2.056
0.941
IRR
Model 5
Table 11.2 Random Effects Negative Binomial Regression of Yearly Economic Protest Events (Weighted by Duration).
oxfordhb-9780199357505.indd 347
10/29/2013 11:10:20 PM
IRR
z-score
Model 1
IRR
z-score
Model 2
IRR
z-score
Model 3
IRR
z-score
Model 4
IRR
z-score
Model 5
Ln population
2.150
5.18**** 2.284
5.28**** 1.987
6.84****
1.969
5.68**** 2.003
7.01****
GDP growth (2006 = 100)
0.922 −4.44**** 0.928 −4.00**** 0.937
−3.91****
0.937
−3.89**** 0.951
−2.85***
Freedom House score, 2006
0.926
−1.15
1.009
0.14
EBRD economic reform index,
2.128
2.95***
2.192
2.39**
2.376
3.73****
1998
Public sector employment (%),
1.049
2.24**
2005
Dummy, IMF standby program
1.836
2.01**
Number of observations
72
72
68
68
72
Number of groups
18
18
17
17
18
Wald chi-square
35.33****
36.20****
51.61****
52.07****
64.08****
Log likelihood
−140.402
−139.747
−130.397
−130.387
−126.623
Variable
Protest Against Economic
Cutbacks
Table 11.2b: Protest Events over Economic Cutbacks
oxfordhb-9780199357505.indd 348
10/29/2013 11:10:20 PM
Coeff.
Coeff.
z-score
Model 2
72
18
55
11.06**
−787.096
z-score
Coeff.
z-score
Model 4
Coeff.
0.67
−0.5
3.02***
−0.34
−2.44**
z-score
Model 5
1.98* 110946.0 2.67***116529.1
−0.82 −3699.1 −0.58 −2099.5
−41059.9 −1.79* −42684.0
0.91 −50970.9 −0.44
4462.5
−54181.0
68
68
72
17
17
18
51
51
55
5.32
9.88**
11.70**
−732.354
−728.499
−786.740
Coeff.
Model 3
1.93* 107930.9 2.91*** 80742.5
−1.21 −1597.5 −0.28 −566.4
−37700.9 −2.45**
83339.4
z-score
Model 1
Ln population
80930.0
GDP growth (2006 = 100)
−7198.3
Freedom House score, 2006
EBRD economic reform index, 1998
Public sector employment (%), 2005
Dummy, IMF standby program
Number of observations
72
Number of groups
18
Uncensored observations (censored at 0) 55
Wald chi-square
4.40
Log likelihood
−789.385
Variable
Participation in All Economic Protest
Table 11.3a: Participation in All Economic Protest Events
Table 11.3 Tobit Regression of Annual Participation in Economic Protest Events.
oxfordhb-9780199357505.indd 349
10/29/2013 11:10:20 PM
Ln population
GDP growth (2006 = 100)
Freedom House score, 2006
EBRD economic reform index, 1998
Public sector employment (%), 2005
Dummy, IMF standby program
Number of observations
Number of groups
Uncensored observations (censored at 0)
Wald chi-square
Log likelihood
Variable
Participation in Protest Against Economic
Cutbacks
72
18
42
9.87***
-583.205
77431.9
−9532.4
Coeff.
Coeff.
z-score
Model 2
72
18
42
14.40***
-581.149
z-score
Coeff.
z-score
Model 4
Coeff.
z-score
Model 5
2.84***82651.9
2.92*** 100347.9 3.76****
−1.96** −7147.9 −1.90* −4564.4 −1.22
−12698.6 −0.85 −32305.8 −2.79***
1.60 43037.5 −0.56
4794.6
1.07
42953.5
0.68
68
68
72
17
17
18
38
38
42
13.22***
13.90***
18.79***
-523.808
-523.451
-566.496
Coeff.
Model 3
2.58***95511.3
3.34*** 72511.2
−2.48**−6948.4 −1.81* −7366.8
−23604.1 −2.15**
88745.8
z-score
Model 1
Table 11.3b: Participation in Protest Events Against Economic Cutbacks
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
350
M ass Polit ics in Tough T imes
growth (relative to 2006) increased the annual number of participants in protests against economic cutbacks by about 9,500 persons, so that an economic
decline of 14 percent below the 2006 level (Latvia in 2009) would be associated with more than 130,000 more participants in protests against economic
cutbacks, whereas a country with 14 percent growth above the 2006 level
(Poland in 2009) would have had 130,000 fewer participants in such protests.
In short, the baseline models show that the effect of economic contraction on
protest mainly occurred through encouraging movement activists to organize
more frequent protests over economic issues (and over economic cutbacks in
particular). But even if we take this into account, the belt-tightening effects of
the recession did encourage greater numbers to participate in protests against
economic cutbacks in those countries that experienced contractions.9
In Models 2, 3, and 4 we introduce two other independent variables that
we expect to affect the level of economic protest in the context of economic
contraction: (1) a country’s Freedom House score (ranging from 2 to 14, with
a score of 2 representing full democracy and 14 representing full autocracy)
for 2006, on the eve of the Great Recession10 ; and (2) the EBRD Economic
Transition Index for 1998, measuring the extent to which a country had made
the transition to capitalist forms of economy in the 1990s.11 We expect a negative relationship between the Freedom House score and economic protest,
as more authoritarian polities should be less tolerant of attempts to organize
protests in general, and individuals living in more autocratic countries should
be less willing to risk participation in the face of negative incentives.
There are also a number of reasons one might expect prior patterns of economic transformation to affect economic protest in the context of the Great
Recession. Those countries that were more progressive in terms of transforming their economies into market economies in the 1990s were the very
countries that earlier had demonstrated “patience” in the midst of economic
contraction. This might have led one to expect similarly quiescent responses
to the Great Recession. However, there are also some reasons to believe the
opposite would be true: those countries that had been in the forefront of economic reform in the 1990s would experience a higher level of protest during
the Great Recession. “Patience” in the 1990s had been closely intertwined
with the promise that rapid economic transition and the additional pain that
countries underwent for the sake of integration with the European Union
would eventually bring a higher standard of living.12 Indeed, the decade
prior to the Great Recession had been one of remarkable economic growth
throughout the region, reinforcing these expectations. But for many living
in the region, the sharp economic decline of the Great Recession shattered
oxfordhb-9780199357505.indd 350
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
351
those dreams, undermining the implicit economic contract between state and
society on which European integration was built.
It could also be that countries that had reformed themselves more thoroughly in the 1990s were more vulnerable to the effects of the Great Recession
in the late 2000s because of the consequences of economic reform for their
economies. The EBRD Economic Transition Index is strongly correlated
with the proportion of a country’s GDP comprising imports and exports, and
with the role of foreign direct investment in a country’s economy—both factors that would likely make a country more vulnerable to a global economic
downturn. However, these factors would not be expected to operate directly
on protest, but indirectly through their effect on economic decline, so that we
are in essence already controlling for their effects by controlling for the influence of economic growth on protest.
Still, in interpreting the effects of democratization and the transition to
capitalism on economic protest during the Great Recession, we need to proceed with caution; as is well known, there was a close relationship between
democratization and economic reform in the 1990s, and indeed the level of
autocracy in a postcommunist country in 2006 was still strongly negatively
correlated (r = −.73) with the extent to which it had engaged in economic
reform in the 1990s. Therefore, in Models 2 through 4 we test for the effects
of each of these two variables separately and jointly, exploring which provides
better explanatory power, if either. Surprisingly, the findings show that, if we
control for population size and economic growth, a country’s level of democracy had no independent effect on the frequency with which movements
organized economic protests or protests against economic cutbacks. Level
of democratization did, however, have an independent effect on the willingness of people to participate in these protests assuming the availability of an
opportunity to participate. Every one-point increase toward autocracy on the
Freedom House scale resulted in about 38,000 fewer protestors per year over
economic issues in general and about 24,000 fewer protestors against economic cutbacks. Thus a country such as Russia, which is relatively high on the
Freedom House scale (i.e., more autocratic) would be expected to have about
340,000 fewer participants in economic protests in general during this period
and 216,000 fewer participants in protests against economic cutbacks than a
country such as Latvia, which is relatively low (i.e., more democratic) on the
Freedom House scale.
By contrast, we found that the EBRD Economic Transition Index, despite
having no relationship with the frequency of economic protest in general
or with participation in economic protests or in protests against economic
oxfordhb-9780199357505.indd 351
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
352
M ass Polit ics in Tough T imes
cutbacks, was strongly associated with a significant increase in the frequency
of protest against economic cutbacks (with every point on the EBRD index
more than doubling the rate of protest on these issues, controlling for the
effects of population size and economic growth). Thus, one would expect three
times as many protests over economic cutbacks in a country such as Hungary
(on the cutting edge of economic reform in the 1990s) than in a country such
as Ukraine (a perennial laggard), even though the two countries experienced
similar levels of economic contraction during the Great Recession. In short,
the very countries that had led the transition from socialism to the market in
the 1990s with relatively little protest were in the forefront of protest over the
effects of the Great Recession in the late 2000s.
Finally, in Model 5 we add two additional factors that might also be associated with economic protest in the midst of economic decline: (1) public
sector employment as a percentage of total employment, and (2) whether
a country adopted an IMF Standby Arrangement in order to deal with the
financial fallout from the recession. High public employment could affect
economic protest during recession in several ways. Countries with large
public bureaucracies that had failed to contain public spending were highly
exposed to the impact of the financial collapse at a time when funds no longer flowed as easily into public coffers. Public employees and their benefits
are likely to be the first targets of cuts when governments are forced to rein
in spending during difficult times. Moreover, public employees tend to be
highly unionized and organized, and therefore relatively more easily mobilized in the face of threats to their standard of living. Highly visible and easily politicized as both domestic political failure and foreign interference in
domestic affairs, IMF rescue arrangements have often served as pretexts for
protest waves around the world, as they typically require the adoption of austerity measures aimed at bringing the borrower’s economy back into balance.
The IMF played a major role in extending loans to postcommunist countries
during the transition from socialism without eliciting much protest reaction.
However, if the Great Recession brought a climate of diminished “patience,”
we might expect that the effect of IMF rescue packages in postcommunist
Eastern Europe would be no different in the rest of the world, where they are
often associated with protest.
As Model 5 indicates, every additional percentage point of total employment that resided in the public sector increased the frequency of protest
against economic cutbacks by about 5 percent. Thus, a country such as Latvia
(which had a proportion of public sector employment of almost 35 percent)
would be expected to experience nearly 50 percent more protests against
oxfordhb-9780199357505.indd 352
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
353
economic cutbacks compared to Estonia (whose public sector employment
was only 24.5 percent of total employment). The adoption of an IMF Standby
Arrangement increased protest against economic cutbacks by 84 percent.
Those postcommunist countries that experienced the highest levels of economic protest and violence associated with the Great Recession were indeed
those that relied on IMF loans to prop them up, and they were also early
seekers of entrance into the EU (Hungary, Latvia, and Romania). By contrast,
postcommunist countries that borrowed heavily from the IMF during this
period but were not EU members (Belarus and Ukraine) experienced little
protest associated with their adoption of Standby Arrangements.13 Neither
public employment nor IMF loans had any independent effect on economic
protest overall or on level of protest participation.
These findings held up to a number of robustness checks.14 Next we explore
them in more detail through two paired case studies (Estonia-Latvia and
Hungary-Ukraine) that illustrate some of the causal mechanisms involved
and identify a number of additional processes that are not easily incorporated
within a quantitative framework.
Latvia and Estonia: A Paired Case Comparison
Latvia and Estonia have similar economic and political reform records; both
experienced similar EU integration processes; their economies have similar
sectoral structures; both enjoyed buoyant pre-crisis economic growth,15 followed by severe economic contractions in 2008 and sharp rises in unemployment in 2008–09. In light of these similarities, the stark contrast between
Latvia’s high number of economic protests and participation rates and
Estonia’s quiescence throws into sharper relief more fine-grained differences
between the two countries that help to explain these differing protest outcomes. Latvia’s higher level of public sector employment (34.7 percent in
2005, compared to 24.5 percent in Estonia)16 and the IMF rescue package in
Latvia are obvious differences, but other factors contributed to the contrasting patterns of protest as well.
One might have expected both Latvia and Estonia to exhibit significant
protest mobilizations in the early 1990s. Protest for independence in the late
Soviet era had created a powerful precedent, and the economic reform process was radical and imposed significant social costs. However, the absence
of protests organized by ethnic Estonians and Latvians may be at least partly
explained by the shared drive to build an independent state and “return
to Europe.” For their part, the sizeable Russophone populations in both
oxfordhb-9780199357505.indd 353
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
354
M ass Polit ics in Tough T imes
countries were initially politically disenfranchised and encountered persistent obstacles to full social and political integration. Moreover, Russophones
are a diverse group with a shared language rather than clear social or ethnic
markers; many had opportunities to migrate to Russia or work in Europe;
Russophone parties lacked organizational coherence, and the mainstream
political discourse was skewed to the center-right and right, tending to crowd
out minority views. These factors may go some way toward explaining the
relative quiescence of Russophone populations in both Latvia and Estonia.
In the absence of systematic data on the participation of specific ethnic
groups in economic protests during the Great Recession, we cannot address
this cleavage systematically. However, very few economic protests in Estonia or
Latvia during the Great Recession seem to have had a distinct ethno-linguistic
dimension to them, at least as reported in the media. Public sector employment is higher among ethnic Latvians, and outmigration (within the EU) has
been higher among the Russophones, suggesting that ethnic Latvians might
have accounted for a significant (if not the main) share of economic protesters
during the crisis. In Estonia, neither Estonians nor Russophones mobilized
over economic conditions during the Great Recession in significant numbers.
Although Estonia and Latvia tend to be discussed as variations of the
same case in terms of economic and political reforms, a closer look reveals
important differences that shaped how they experienced the boom and bust
cycle of the 2000s and the strict austerity measures both countries imposed
on their populations. One crucial difference is that Estonia consistently pursued a balanced budget and generated a budget surplus since 2002, which it
used to build up a Stabilization Reserve Fund precisely for emergency situations (Kraan, Wehner, and Richter 2008, 10; Martin 2010). By contrast, Latvia
did not secure its reserves but instead spent its growth proceeds (Kraan et al.
2009, 190). As a result, Estonia’s austerity measures could concentrate on cutting other expenditures than social transfers, whereas Latvia’s IMF-overseen
austerity program introduced cuts across the board. By softening the blow of
austerity cuts, Estonia’s balanced budget and financial reserves helped the government maintain a public reputation for effectiveness. (On the importance of
budget deficits for protest mobilization, see Kriesi’s Chapter 10 in this volume.)
Similarly, different degrees of corruption led to variation in levels of public trust in government in the two countries. Indeed, on a number of corruption measures Latvia has performed consistently worse than Estonia.17 Latvia
entered the financial crisis with the issue of corruption conspicuous on the
public agenda after a number of high-level corruption allegations against
party and government officials and controversial government decisions aimed
oxfordhb-9780199357505.indd 354
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
355
at restraining its Anti-Corruption Bureau, which had been investigating several prominent businessmen with political influence. Several demonstrations
over these issues had mobilized sections of society already prior to the onset
of the Great Recession (in particular, a rally in October 2007, at which several thousand protesters called for the resignation of Prime Minister Aigars
Kalvitis of the People’s Party over his decision to dismiss the head of the
Anti-Corruption Bureau) and opened the way for a series of protests driven
by economic demands once the implications of the crisis became more tangible. The prime minister was forced to step down in early December 2007
after President Zatlers called for his resignation. Corruption issues remained
high on the political agenda, with the chief of the Anti-Corruption Bureau
eventually dismissed in June 2008 in yet another controversial decision, leaving the bureau without a new head until early 2009.
Estonia’s transition from communism has been less conflictual than
Latvia’s, and the quality of formal and informal institutions has been higher
(Kuokstis and Vilpisauskas 2010; Pettai and Mölder 2010). Ironically, when
the Latvian Prime Minister Kalvitis resigned in late 2007 in the wake of corruption scandals after three years in office, he had become the longest-serving
prime minister in Latvian politics since 1991. The political volatility in Latvia
(compared to government stability in Estonia) underpinned perceptions of
a lack of government effectiveness. This volatility was only partially masked
by a repeated pattern of keeping the government in place while replacing its
members (a pattern that continued during the financial crisis).
The link between perceptions of government ineffectiveness or corruption
and protest is an indirect but important one. These perceptions undermined
public trust in government, which in turn helped to politicize society in the
context of a crisis situation. In Latvia public trust in the government was low
(20 percent in spring 2007, and still only 13 percent by spring 2010). By contrast, even in the midst of severe economic contraction it remained high in
Estonia (66 percent in spring 2007, and 53 percent in spring 2010)—significantly higher than the EU average of 41 percent for 2007 and 29 percent for
2010 (Eurobarometer 67, 2007; Eurobarometer 73, 2010).
On the eve of the Great Recession, recognition of past economic progress and
expectations of continued economic improvement were high in the Baltic states.
When asked in spring 2007 if their situation had changed in the last five years,
75 percent of Estonians said their situation had improved, 16 percent said it was
unchanged, and only 8 percent thought their situation had worsened. Latvians
were somewhat less positive but still well above the EU average at the time: 58 percent thought their situation had improved, 23 percent saw it as unchanged, and
oxfordhb-9780199357505.indd 355
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
356
M ass Polit ics in Tough T imes
18 percent said it had worsened (Eurobarometer 67, 2007, 265). The outlook for
the next five years was similarly optimistic: 66 percent of Estonians and 58 percent
of Latvians expected their situation to improve in the next five years (well above
the 43 percent average for the EU as a whole; Eurobarometer 67, 2007, 266). The
impact of the crisis would have thus come as a shock to populations in both countries, but Estonia’s policy track record and its stable and trusted government provided it with a buffer that was lacking in Latvia.
Latvia’s large public sector also played an important role in spurring protest in response to the economic crisis. In September 2008 health workers,
teachers, and police demonstrated in Riga to demand higher salaries (Telegraf,
September 19, 2010). Soon afterward, on October 30, about a thousand doctors and nurses began a two-day strike while the Saeima (the Latvian legislature) was debating an austerity budget envisaging job and health expenditure
cuts. In December the Latvian government agreed to an IMF rescue deal
(involving the EU) of €7.5 billion (of which €600 million was immediately
disbursed, with further tranches earmarked for 2009–10). As a result, a range
of austerity measures were put in place to bring the budget deficit under control, including job cuts and an average public salary cut of 25 percent. The
IMF deal proved the crucial catalyst for mobilizing the public sector at large.
On January 13, 2009, large-scale antigovernment protests involving about
ten thousand people turned violent when hundreds of youths (apparently a
mixture of Russian speakers and Latvians) overturned cars, looted shops, and
threw cobblestones ripped from the streets of the old city into the windows
of Saeima’s building (Reuters, January 13, 2009). This was the worst violence
in Latvia since 1991, and it triggered a swift political response by President
Zatlers, who openly criticized the coalition government for having lost touch
with voters. Following the Trade Union Federation’s call for a constitutional
referendum, he asked the Saeima to strengthen his constitutional powers to
initiate new elections in the event of legislative gridlock and to give the public
the right to seek early elections via a referendum. Zatlers also urged the government to facilitate a turnover in government posts and bring opposition parties
into the cabinet (Reuters News, January 14, 2009). Prime Minister Godmanis
survived a vote of no confidence in early February, but his position became
untenable when two of his coalition partners left the government. In March
2009, Valdis Dombrovskis of the New Era Party took over as prime minister.
Throughout 2009—in the run-up to and after the Saeima’s approval of further
budget cuts to qualify for a tranche of the IMF package (AP, June 18, 2009)—
smaller and medium-size anti-austerity protests, led by public sector workers,
continued. Dombrovskis temporarily oversaw a minority government, but
oxfordhb-9780199357505.indd 356
10/29/2013 11:10:21 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
357
he won a narrowly contested election in October 2011 against the Harmony
Center, the center-left party associated with the Russophone electorate. Early
parliamentary elections held in September 2011, after a referendum on the dissolution of parliament, kept Dombrovskis in office as part of a coalition government, although his party had actually come in third, after Harmony Center
and the new Reform Party set up by former President Zatlers.
Estonia’s government stability and relative quiescence in the midst of an
analogous economic decline make for a marked contrast. Prime Minister
Andrus Ansip (Reform Party) was appointed in 2005 to replace Juhan Parts
and managed to see out an entire elected term in office from 2007, surviving a
vote of no-confidence over austerity measures in February 2009 and overseeing a minority government after his coalition partner, the Social Democrats,
left in May 2009. He was reelected (albeit narrowly) in March 2011, again as
part of a coalition of center-right parties (with Pro Patria and Res Publica
Union; European Voice, February 24, 2011).
Estonia’s entrance into the eurozone on January 1, 2011, also set Estonia apart
from Latvia (although Latvia maintained its currency peg against the euro).
After EU membership, the adoption of the euro had become the next big international test of new member states’ economic and financial credentials. Estonia’s
commitment to staying on target for euro membership despite the financial crisis provided an external and domestic anchor and reinforced its image of policy
coherence. Because Estonia did not need to be bailed out with the help of an
IMF loan, the government could present itself as strong, providing an additional
societal buffer against protests in the midst of severe economic pain.
The comparison between Latvia and Estonia confirms that a large public
sector hit particularly hard by austerity measures played an important role
in evoking economic protest, and that IMF rescue packages also acted as
an important lightning rod for protest. The comparison also brings out the
importance of domestic differences related to the public’s trust in transparent and effective government, the existence of financial reserves, and external
reform anchors (such as the adoption of the euro) for explaining differential
protest responses to the economic disaster.
Hungary and Ukraine: A Paired Case
Comparison
Hungary and Ukraine varied greatly in the type and comprehensiveness of
their economic and political reforms in the 1990s,18 but in the middle to late
oxfordhb-9780199357505.indd 357
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
358
M ass Polit ics in Tough T imes
2000s both countries experienced a deep financial and economic crisis that
required the help of the IMF. In Hungary a severe domestic crisis after many
years of expectations tied to the country’s reform success was reinforced by
the global financial crisis; in Ukraine external shocks abruptly ended a short
period of strong economic growth and plunged the country back into its
long accustomed crisis. Hungary saw sustained large-scale protest mobilization, whereas Ukraine continued to experience only small-scale protests on a
wide range of economic and political issues. These differing trajectories and
Hungary’s larger public sector (31.5 percent in 2005, compared to 21.6 percent
in Ukraine) go some way toward explaining why protest activity was more
significant in Hungary than in Ukraine. But the disparate political landscapes
of these countries also provide some critical clues to the dynamics involved
in producing differing protest responses to the Great Recession. Indeed, they
point to the fact that economic conditions, a sizeable public sector, and a
catalyst such as an IMF loan and accompanying austerity programs need to
be activated by social movements or political parties in order to translate
into significant protest. Hungary and Ukraine share an extremely polarized
political environment—arguably a conducive starting-point for protest. But
they are polarized differently, and the nature of this polarization, combined
with differing expectations within their respective publics about government,
played important roles in producing contrasting protest outcomes. Hungary
belonged to the frontrunners of reform in the region, and after more than two
decades since the start of comprehensive economic reforms and several years
into EU membership, a domestic financial crisis and the immediate austerity
program hit society hard, further polarized it, and fed into party-led mobilization. Ukraine’s extremely drawn out political and economic reform process
since 1991 and the disillusionment that enveloped Ukrainian society after the
enormous mass mobilizations of the Orange Revolution in 2004 made the
new financial and economic crisis less of a rallying point for political parties
and for society at large.
In 2006 Hungary had the largest budget deficit in the EU. Austerity
measures, including an envisaged 10 percent cut to the public sector, were
introduced before the onset of the Great Recession in response to the country’s economic problems. The domestic economic crisis unfolded against
an extreme and growing political polarization between the Socialist Party
(MSZP) and the center-right party Fidesz under Viktor Orbán. A leaked
“secret speech” by Socialist Prime Minister Ferenc Gyurcsány in autumn 2006
in which he admitted that he had lied about the state of the economy before
the elections provided the trigger for a long series of antigovernment protests,
oxfordhb-9780199357505.indd 358
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
359
culminating in Fidesz’s landslide victory in the April 2010 elections, when the
political center/center-left was practically obliterated and democratic checks
and balances came unhinged.
From early 2007, there were repeated antigovernment protests of varying magnitude (several hundred to several thousand) in Budapest and cities
across Hungary demanding the prime minister’s resignation over the government’s plan to raise taxes, cut subsidies, and dismiss thousands of public sector employees (AP, January 20, 2007). In February antigovernment protesters
staged road closures in a hundred locations across the country in protest
against the austerity package. In addition, there were sector-specific protests
and local issue-specific protests tied to the austerity measures.
During these first protests Fidesz still appeared “weak and divided”
(New York Times, March 11, 2007). However, Fidesz gradually stepped up the
pressure on the coalition of Socialists and liberal Free Democrats (SZDSZ)
by actively shaping the public discourse and organizing protests (some deliberately coinciding with national holidays), using the social unrest over austerity measures to boost its own electoral support. Antigovernment protests
increasingly acquired nationalist overtones, as Fidesz faced growing competition from the extreme right in Jobbik, a right-wing youth movement transformed into a party in 2003 that was gradually increasing its support base.
In September 2007 antigovernment protests marked the first anniversary of
the leaked prime minister’s speech, with mass protests against anti-austerity
measures mobilizing an estimated ten thousand people.
In late October and early November, train drivers staged warning strikes
across the country against government plans to close various rail routes. This
was followed by about half of Budapest’s bus drivers and about four thousand bus drivers across Hungary, who protested against planned layoffs. In
March 2008 a Fidesz-initiated referendum, approved by the Constitutional
Court, put three questions to the public: on the introduction of doctors’ fees,
hospital charges, and university tuition. More than 80 percent of those who
participated voted against all of the proposals.
Hungary narrowly averted a financial crisis with the help of an
International Monetary Fund–led $25.1 billion rescue package put in place in
November 2008. It involved the EU and was tied to strict austerity measures
aimed at controlling the budget deficit. Hungary’s currency fell to an all-time
low against the euro, as investors fled amidst concerns over the country’s high
budget deficit, public sector debt, and heavy reliance on external financing
(Reuters, November 29, 2008). Large-scale foreign borrowing had allowed
the country to run a sizeable current account deficit (7.5 percent of GDP in
oxfordhb-9780199357505.indd 359
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
360
M ass Polit ics in Tough T imes
2006, and 6.4 percent in 2007) and a large fiscal deficit in 2006 (Cordero
2009). Access to cheap international credit saw the private sector and
households exposed to high debts, often held in foreign currencies. When
external capital inflows dried up in connection with the global financial crisis, the Hungarian economy imploded. The presence of the IMF mission
further fueled the ongoing protests. In late November 2008 they built into
large-scale anti-government actions against pay and pension cuts, with about
ten thousand workers across Hungary going on strike at train stations, manufacturing plants, schools, pharmacies, and electricity plants. Several thousand
public sector workers represented by thirty unions also protested outside
the Hungarian National Assembly against pay cuts (Reuters, November 28,
2008). On December 15 the budget tabled by the prime minister, including a
slight upward revision of the fiscal deficit targets that was sanctioned by the
IMF, passed the National Assembly. Immediately, rail worker strikes started
up again, and public sector unions announced a new series of strikes for early
2009 (Reuters News, December 15, 2008).
The momentum of regular Sunday demonstrations against the government
gradually picked up throughout the spring of 2009. In March Prime Minister
Gyurcsány resigned in the hope of enabling the Socialist Party to catch up in
the polls before the elections scheduled for early 2010. In mid-April Gordon
Bajnai (MSZP) was sworn in as the new prime minister. In Hungary’s highly
politicized setting, this internal turnover triggered further demonstrations,
strikes, and clashes with the police. The extremist party Jobbik achieved a
breakthrough in the European Parliament Elections in June 2009 (gaining
three representatives) and continued to position itself in anticipation of the
2010 elections. Large-scale anti-austerity protests by public sector workers
continued until the elections in April 2010, which saw Fidesz win a landslide
victory and Viktor Orbán become prime minister, commanding a two-thirds
(constitutional) majority in the National Assembly without an effective
opposition. Jobbik gained parliamentary representation (twenty-six seats) as
Hungary’s third-strongest party. It had played an important role in the mobilization of sustained protest and affected a significant shift of Hungary’s party
system to the right.
Compared to Hungary, Ukraine’s transition in the 1990s proved considerably more complex. This was due to the country’s integration into the
Soviet economy, the need to build a new independent state alongside fundamental political and economic reforms, and the existence of strong regional
cleavages not only around language and identity issues but also including
differences in economic structures and domestic and international political
oxfordhb-9780199357505.indd 360
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
361
preferences (Sasse 2010). In addition to the post-Soviet output collapse, soft
budget constraints saw the country slide into hyperinflation by late 1993 and
only gradually recover through monetary stabilization in 1996. The privatization process was slow and gathered speed only in the late 1990s (D’Anieri,
Krawchuk, and Kuzio 1999, 166–205). And it was not until the early to
mid-2000s that Ukraine began to experience significant positive economic
growth. Presidential policy wavered continually during these years to appease
certain regional constituencies, numerous governments collapsed, and the
constitutional division of power between the president and the prime minister became politicized (and remains ambiguous even today).
The authoritarian drift of the scandal-ridden Kuchma regime ended
abruptly with the Orange Revolution in late 2004, a moment of massive protest mobilizations (centered on Kyiv and the western regions) against electoral
fraud in the presidential elections. The protests led to a rerun of the elections
and brought to power a reform coalition of President Viktor Yushchenko
and Prime Minister Yuliya Tymoshenko. After many years of economic crisis, corruption, and an uncertain foreign policy course, public expectations
for domestic reform and Ukraine’s European integration were high, at least
in western and central Ukraine. However, these expectations were soon
dashed by the growing policy disagreements and personal rivalry between
Yushchenko and Tymoshenko and by repeated government changes that saw
the return of Yushchenko’s 2004 electoral rival Viktor Yanukovych as prime
minister in 2006 (and eventually, in 2010, as president). Corruption remained
rampant, further undermining societal trust in government.19 Opinion polls
showed that throughout 2007 around 40 percent of respondents did not support the activities of the government; this proportion rose throughout 2008,
peaking in March 2009 at 63.7 percent (Opinion Polls for 2007–2009, from
the Razumkov Centre). Even after the presidential elections in 2010 that were
won by Yanukovych, 30 percent immediately disapproved of his new government, and opposition steadily rose to 45 percent by the end of 2010 and
to 62.6 percent in 2011 (Razumkov Centre, Opinion Polls, 2010–2011). In
response to the question of whether Ukraine was developing in the right or
wrong direction, there was only a brief period in early 2005, immediately after
the Orange Revolution, when more than 50 percent of Ukrainians responded
positively, dropping to about 20 percent by the end of 2007, and even lower
by 2009 (Razumkov Centre, Public Opinion, 2004–2008). Such widespread
disillusionment with Ukrainian politics that cut across the political spectrum
and across the country’s regional divide limited the ability of any political
actor to mobilize public support around a tangible political alternative. It
oxfordhb-9780199357505.indd 361
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
362
M ass Polit ics in Tough T imes
would be misleading to characterize Ukrainian society as apolitical; there was
persistent low-level protest mobilization (at times reaching several thousand
protesters) in Kyiv and regional cities over a wide range of political issues,
such as opposition to NATO membership, Crimean Tatar mobilization, protest over economic issues, and pre-election rallies orchestrated by the two
main parties (Yanukovych’s Party of Regions and Tymoshenko’s Bloc). But
these scattered protests have not jelled into larger campaigns or events. None
of the dominant political actors had the credibility or vision to formulate a
tangible political alternative that would tap into the alienation accompanying
the economic crisis, not even for parts of the country.
Ukraine’s brief period of impressive economic growth in the mid-2000s
(GDP grew by more than 12 percent in 2004 and by more than 7 percent
in 2006 and 2007) was driven by the steel sector benefiting from high
global steel prices and was accompanied by property and credit booms. All
this came to a halt as a result of the external shock of the global financial
crisis. However, even the IMF rescue package of $16.4 billion, agreed on in
late October 2008, sparked only a few medium-sized protests and no sustained protest campaigns. The last installment of the IMF loan was frozen
in late 2009 when the Ukrainian government raised minimum wages and
pensions. The involvement of the IMF—against a backdrop of earlier IMF
loans in the late 1990s—failed to act as a catalyst for large-scale protests like
those in Latvia or Hungary. Ukraine’s biggest anti-austerity protests (up to
five thousand protesters) occurred in December 2008; the implications of
the IMF deal for jobs and social transfers were singled out by the protesters (mostly coal miners from state-run mines and pensioners) at the time
(DPA, December 23, 2008). In February 2009 several thousand people protested against the austerity budget of the city of Kyiv. The protests brought
together representatives of the Ukrainian Trade Union Federation, workers
of state-run companies, public sector employees, and political activists from
both Yanukovych’s and Tymoshenko’s parties (Ukraine General Newswire,
February 12, 2009). The biggest protest since the legislative elections in 2007
took place in April 2009—estimates vary between twenty thousand and fifty
thousand participants—at a pre-election rally by Yanukovych supporters
demanding the resignation of the president and prime minister for their failure to resolve the economic crisis (Reuters, April 3, 2009; dpa, April 3, 2009).
This event occurred a week after the legislature passed two bills restoring the
flow of IMF credits. In July 2010 a new IMF loan of $15.2 billion was agreed
on, directly linked to a significant hike in household gas and utility prices. The
period October–December 2010 also saw a series of medium-sized protests
oxfordhb-9780199357505.indd 362
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
363
(involving several thousand people) by small business owners and employees
against the new tax code, resulting in President Yanukovych vetoing the provisions on higher taxes for small businesses (AFP, November 30, 2010). Though
effective on their own terms, the interests represented were issue-specific and
failed to mobilize a wider societal concern about the economic crisis.
For a country the size of Ukraine and with the depth of economic contraction that Ukraine experienced during the Great Recession, economic protest
was decidedly small. The Ukrainian case illustrates that economic crisis, an
IMF program, the precedent of recent massive mobilization, and a high level
of political polarization need not translate into large-scale mass protest without parties mobilizing along this cleavage and without the public believing in
the prospect of a credible alternative. In Ukraine’s postrevolutionary political
landscape, disillusionment with all the key actors prevails. A series of elections in close succession provided sufficient opportunities for the main political parties to mobilize around the fallout from the economic crisis; however,
they chose instead to tap into familiar regional identity cleavages without
extensively activating the issues of profound economic hurt that Ukraine has
experienced as a result of the Great Recession. By contrast, in Hungary the
political landscape was polarized between left- and right-wing parties. Given
that the homegrown financial crisis prior to the Great Recession was associated with the Socialist Party, the combined effect of domestic and international crises gave the right-wing party Fidesz an opportunity to stoke and
channel public discontent into electoral support. Growing support for the
extreme right party Jobbik moved Fidesz further to the right. Ukraine’s political scene is similarly polarized between two political parties and their leaders
(Yanukovych and Tymoshenko), but they represent regional constituencies
rather than clear ideological programs. Having rotated in and out of office,
both leaders failed to articulate a credible alternative amidst the widespread
disillusionment of Ukrainians across regional and political affiliations.
Conclusion
It is hardly surprising to find that in the region of the world hit hardest by
the Great Recession in the 2007–2010 period, significant outbursts of protest occurred. However, the story of the end of “patience” in Eastern Europe
is not quite that simple. As we have seen, countries in the region weathered the economic crisis differentially, with some continuing to experience
economic growth and others undergoing severe contraction. In Tolstoyan
fashion, those “happy” countries that continued to experience economic
oxfordhb-9780199357505.indd 363
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
364
M ass Polit ics in Tough T imes
growth in the midst of global crisis were all little affected by protest, while
those “unhappy” countries that experienced significant economic contractions were all “unhappy” in their own ways, displaying quite varied protest responses to economic decline. We showed that those countries that
were most vulnerable to a high level of economic protest during the Great
Recession were precisely those that had been in the vanguard of economic
and political reform in the 1990s, in part because they were more vulnerable
to economic downturn as a result of their high dependence on the global
economy, in part because they had generated huge expectations surrounding
their track records of reform and EU integration, which were dashed in the
context of the Great Recession. But a number of other factors also shaped
differential protest responses to economic crisis. Countries with high levels
of public sector employment experienced greater protest, as austerity measures targeted public employees in particular, and public employee unions
provided the mobilizing capacity for generating large and sustained protest
campaigns. IMF rescue packages yielded both an opportunity and a grievance for accelerating protest, through the cutbacks they mandated and in
how they demonstrated government incapacity. Governments that, prior to
the onset of the crisis, already lacked public trust or a reputation for effectiveness were more likely to see sustained and significant protest. We also
showed the importance of political context and the key role that political
parties and public expectations play in politicizing the cleavages born out
of economic crisis. (For further analysis of the link between electoral politics and protest mobilization in response to the crisis, see Kriesi’s Chapter 10
in this volume.) If parties lack the credibility or desire to mobilize citizens
around the issues of economic pain that they are experiencing, and if citizens
are disillusioned and incapable of imagining that collective action might be
an effective remedy to their plight, then even a severe economic contraction,
an IMF austerity program, and a past history of mobilization will not produce significant or sustained waves of protest.
Thus the protest response to the Great Recession in postcommunist
East Europe shows that there is certainly a relationship between deprivation
and protest. But it also shows that this relationship is hardly a simple one
and depends on the presence of particular facilitating structural conditions,
opportunities, and mobilizing structures. The stark contrast between the relatively more “patient” or quiescent protest responses in Eastern Europe during
the transition to capitalism in the 1990s and the more contentious protest
responses to the Great Recession (even though they both involved deep
economic pain) raises broader questions about the conditions under which
oxfordhb-9780199357505.indd 364
10/29/2013 11:10:22 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
365
individuals mobilize collectively against economic hardship. “Patience,” it
seems, is much harder to sustain the second time around.
Appendix: Procedures for Coding of Event Data
For the purposes of this study, a demonstration was defined as an event (1) that was a voluntary gathering of persons with the purpose of engaging in a public collective display
of sentiment for or against public policies, (2) that involved a minimum of one hundred
persons, and (3) in which the number of participants was not restricted by the organizers of the event (i.e., it was not a conference, convention, or other restricted organized
meeting). A strike was defined as a work stoppage with the aim of protesting the policies of government, enterprise management, or some other authority (with a minimum
size of one hundred participants). For both demonstrations and strikes, a minimum
of one hundred participants was required for an event to be included in the sample,
since events with fewer participants than that are usually not reported in media sources.
A mass violent event was defined as collective violence against people or property, with
a minimum size of fifteen participants. For ease of coding, any protest act that lasted for
a portion of a day was counted as having lasted for an entire day.
Searches of Reuters were conducted using Factiva, while searches of AP, Agence
France Presse, Deutsche Press Agentur, and Interfax were conducted using LexisNexis.
To ensure that the use of index terms and relevancy filters on searches undertaken in
LexisNexis did not bias the results, the list of article headlines returned using the narrow search criteria was systematically compared for a period of six months with the list
of article headlines generated by an unfiltered, traditional subject search. No biases were
found. As a further check on the coverage of the five international wire services, searches
of national newswire services for seventeen of the eighteen countries were undertaken
for a period of three months to see whether local coverage of protest events differed systematically from international coverage. The results suggested that coverage of events
in Hungary, the Czech Republic, and Lithuania by the international newswires likely
underestimated the extent of politically salient protest in these countries. Between
January 1 and March 31, 2009, national newswires reported on nine events in Hungary,
eight events in the Czech Republic, and seven events in Lithuania that were not covered
by the five international wire services. In all other cases, however, coverage of events was
roughly identical. At the same time, the results suggested that national wire services
had their own substantial lapses in coverage, as a number of the events captured by the
international newswires went unreported by the national outlets examined.
Once the media reports were collected, they were examined in detail by one of three
members of the coding team and coded using a common coding protocol and detailed
set of instructions into a relational database. Information on fifty-five fields was collected, including data on the location of the event, the type of event, the duration of
the event, the number of participants, the specific demands made and the targets of
protest, the types of participants and the organizers of protest, the police response and
oxfordhb-9780199357505.indd 365
10/29/2013 11:10:23 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
366
M ass Polit ics in Tough T imes
the number of arrested or injured, and the extent of civilian-led violence and the degree
of damage it caused. All sources used to code each record were hyperlinked to the individual record, thereby allowing the use of this rich body of reports on protest activity
as part of the parallel qualitative analysis in the second half of this essay. In order to
ensure inter-coder reliability, coders met weekly to resolve questions jointly (and with
participation of the principal investigators). A sample check on inter-coder reliability
for a sixth of the coded cases revealed an inter-coder reliability of approximately 95 percent. The coding of all events was also reviewed by the principal investigators to further
ensure accuracy and consistency.
Notes
1. Data comes from Beissinger (2002).
2. In our sample, the duration of protest events varied from a portion of a day to a
high of 134 days, with more than 90 percent of events lasting for a day or less.
3. The number of participants in the events we recorded ranged from a minimum of
100 to approximately 800,000, with a median of 738.
4. In the scholarly literature on cross-national patterns of protest, population size is
normally related to the amount of protest in a society (for example, Hibbs 1973,
25). This well-researched pattern accords with the assumptions of critical mass
theory, which argues that the larger and more variegated a population, the more
likely it is that groups will find a sufficient number of participants to overcome
collective action problems (Marwell and Oliver 1993). Indeed, as we will show,
across the eighteen postcommunist countries examined here population size was
systematically related both to the frequency of protest events and to the level of
participation in them.
5. Property damage was coded as follows: those events that involved no violence
were coded as 0; those involving violence but no property damage as 1; those
involving minor property damage as 2; those involving sporadic looting or arson
as 3; those involving widespread looting or arson as 4; and those involving major
property destruction as 5.
6. We use event counts weighted by duration, as is common practice in event-count
analyses, since this better reflects protest effort than simple event counts.
7. Since this is a short panel (only four years), and some of the independent variables
to be tested are time-invariant, a fixed-effects model would not be appropriate.
See Cameron and Trivedi (1998, 287–292). For robustness checks, we tested these
results against a pooled sample with standard errors adjusted for clustering by
country.
8. On cross-sectional time-series tobit models, see Arellano and Honoré (1998).
9. For all four models we tested for the presence of any lagged effects of economic
growth on protest and found none.
oxfordhb-9780199357505.indd 366
10/29/2013 11:10:23 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
367
10. By using 2006 (the year prior to our first measurements of protest), we seek to
avoid potential issues of endogeneity, as protest itself (and reactions to it on the
part of regimes) could affect the measurement of democracy.
11. The EBRD Transition Index includes thirteen measures of economic reform on
issues of privatization, enterprise governance and restructuring, price liberalization, trade and foreign exchange, competition policy, banking reform, securities
markets, and infrastructure reform, each ranging in score from 1 (little reform) to
4.33 (the level typical of advanced industrial economies). The measure used here
represents an average of the thirteen measures. Data for the Czech Republic are
missing from the EBRD data.
12. On the stereotypes about the West associated with EU accession among
Hungarians, for instance, see Fölsz and Tóka (2006).
13. Essentially, this factor is controlled for in the protest event models in Table 11.2
by the EBRD Economic Transition Index, which is highly correlated with EU
membership.
14. We tested the event models and the tobit models using a pooled sample with standard errors adjusted for clustering at the country level, with no major changes in the
substantive results. (The pooled model for economic protest in general, however, did
indicate very weak statistical significance for the EBRD Transition Index and IMF
Standby Arrangements.) We also tested for the inclusion of a series of other independent variables to see if they changed the results in any way (with no effect): the degree
of fractionalization of government parties in the legislature, the degree of fractionalization of opposition parties in the legislature, the seat difference between the largest
government party and the largest opposition party in the legislature, the difference
in vote share between government parties and opposition parties, the percentage of
unemployed, and gross debt as a percentage of GDP. We did find some additional
effects from a yearly change in unemployment variable, which in the event count
models soaked up much of the effect of our GDP growth variable (as one might
expect) but had no effect on the findings overall or on their interpretation.
15. According to the World Bank (2006), Latvia recorded the fastest economic
growth in the EU in 2006 (12.2 percent GDP growth), with Estonia not far
behind (10.6 percent).
16. Figures are computed from EBRD Structural and Institutional Change Indicators
data, using the share of the private sector in total employment in order to estimate the share of public sector employment as a percentage of total employment.
The data are available at http://www.ebrd.com/pages/research/economics/data/
macro.shtml.
17. According to Transparency International’s Corruption Perceptions Index (CPI),
ranking more than two hundred countries worldwide, Latvia scored 4.8 (out
of 10) in 2007 and came in thirty-ninth out of the countries included, compared to Estonia scoring 6.5 and being ranked twenty-eighth. See Transparency
International, CPI (2007).
oxfordhb-9780199357505.indd 367
10/29/2013 11:10:23 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
368
M ass Polit ics in Tough T imes
18. Hungary had an EBRD economic reform index of 3.92—significantly higher than
Estonia and Latvia.
19. According to Transparency International’s CPI, Ukraine scored 2.7 out of 10 in
2007 and ranked 118th out of about 200 countries; see Transparency International,
CPI (2007).
References
Ágh, Attila. 1991. “Transformations to Democracy in East Central
Europe: A Comparative View.” In György Szoboszlai, ed., Democracy and Political
Transformation, 103–122. Budapest: Hungarian Political Science Association.
Arellano, Manuel, and Bo Honoré. 1998. “Panel Data Models: Some Recent
Developments.” In James J. Heckman and Edward Leamer, eds., Handbook of
Econometrics, Vol. 5, 3230–3296. Amsterdam: Elsevier Science.
Beissinger, Mark R. 2002. Nationalist Mobilization and the Collapse of the Soviet State.
Cambridge : Cambridge University Press.
Cameron, A. Colin, and Pravin K. Trivedi. 1998. Regression Analysis of Count Data.
Cambridge : Cambridge University Press.
Cordero, Jose Antonio. 2009. The IMF’s Stand-by Arrangements and the Economic
Downturn in Eastern Europe: The Cases of Hungary, Latvia, and Ukraine.
Washington: Center for Economic and Policy Research, September.
D’Anieri, Paul, Robert Krawchuk, and Taras Kuzio. 1999. Politics and Society in Ukraine.
Boulder, CO : Westview Press.
Ekiert, Grzegorz, and Jan Kubik. 1998. “Contentious Politics in New Democracies: East
Germany, Hungary, Poland, and Slovakia, 1989–93.” World Politics 50 (4), 547–581.
Ekiert, Grzegorz, and Jan Kubik. 1999. Rebellious Civil Society Popular Protest and
Democratic Consolidation in Poland, 1989–1993. Ann Arbor: University of
Michigan Press.
Eurobarometer. 2007. Standard Eurobarometer 63, Spring. http://ec.europa.eu/
public_opinion/archives/eb/eb67/eb67_en.htm.
Eurobarometer. 2010. Standard Eurobarometer 73, Spring. http://ec.europa.eu/
public_opinion/archives/eb/eb73/eb73_en.htm.
Fölsz, Attila, and Gábor Tóka. 2006. “Determinants of Support for EU-Membership in
Hungary.” In Robert Rohrschneider and Stephen Whitefield, eds., Public Opinion,
Party Competition, and European Union in Post-Communist Europe, 145–164.
Houndsmills, Basingstoke : Palgrave Macmillan.
Greskovits, Béla. 1998. The Political Economy of Protest and Patience: East European
and Latin American Transformations Compared. Budapest: Central European
University Press.
Howard, Marc Morjé. 2003. The Weakness of Postcommunist Civil Society.
Cambridge : Cambridge University Press.
oxfordhb-9780199357505.indd 368
10/29/2013 11:10:23 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
An End to “Patience”?
369
Haggard, Stephan, and Robert Kaufman, eds. 1992. The Politics of Economic
Adjustment: International Constraints, Distributive Conflicts, and the State.
Princeton, NJ: Princeton University Press.
Hibbs, Douglas A. 1973. Mass Political Violence. New York: Wiley.
Hibbs, Douglas A. 1976. “Industrial Conflict in Advanced Industrial Societies.”
American Political Science Review 70 (4), 1033–1058.
IMF. 2010. “World Economic Outlook,” January 26, at http://www.imf.org/external/
pubs/ft/weo/2010/update/01/pdf/0110.pdf (accessed on May 17, 2013).
Kraan, Dirk-Jan, Joachim Wehner, and Kirsten Richter. 2008. “Budgeting in
Estonia.” OECD Journal on Budgeting 8 (2), 1–40, http://www.oecd.org/dataoecd/59/23/42007202.pdf.
Kraan, Dirk-Jan, Joachim Wehner, James Sheppard, Valentina Kostyleva, and Barbara
Duzler. 2009. “Budgeting in Latvia.” OECD Journal on Budgeting 9 (3), 1–43.
Kuokstis, Vytautas, and Ramūnas Vilpisauskas. 2010. “Economic Adjustment to the
Crisis in the Baltic States in Comparative Perspective.” Paper presented at the
Seventh Pan-European International Relations Conference, Stockholm, September,
http://stockholm.sgir.eu/uploads/Economic%20Adjustment%20to%20the%20
Crisis%20in%20the%20Baltic%20States%20in%20Comparative%20Perspective.
pdf.
Martin, Reiner. 2010. “Boom and Bust in the Baltic Countries—Lessons to Be Learnt.”
Intereconomics 4, 220–226.
Marwell, Gerald, and Pamela Oliver. 1993. The Critical Mass in Collective Action.
Cambridge : Cambridge University Press.
Ost, David. 2005. The Defeat of Solidarity: Anger and Politics in Postcommunist Europe.
Ithaca, NY: Cornell University Press.
Pettai, Vello, and Martin Mölder. 2010. “Freedom House: Nations in Transit—Estonia.”
June, 29. http://www.unhcr.org/refworld/docid/4d53f0220.html.
Przeworski, Adam. 1991. Democracy and the Market. Political and Economic Reforms in
Eastern Europe and Latin America. Cambridge: Cambridge University Press.
Razumkov Centre. 2004–2011. Public Opinion Polls. http://www.razumkov.org.ua/
eng/poll.php?poll_id=75;
http://www.razumkov.org.ua/eng/poll.php?poll_
id=67; http://www.razumkov.org.ua/eng/poll.php?poll_id=66.
Robertson, Graeme B. 2010. The Politics of Protest in Hybrid Regimes: Managing Dissent
in Post-Communist Russia. Cambridge : Cambridge University Press.
Sasse, Gwendolyn. 2010. “Ukraine: The Role of Regionalism.” Journal of Democracy, 21
(3), 99–106.
Seleny, Anna. 1999. “Old Political Rationalities and New Democracies: Compromise
and Confrontation in Hungary and Poland.” World Politics 51 (4), 484–519.
Tilly, Charles, and Edward Shorter. 1974. Strikes in France, 1830–1968.
Cambridge : Cambridge University Press.
Transparency International. 2006, 2007. Corruption Perceptions Index (CPI), http://
www.transparency.org/policy_research/surveys_indices/cpi.
oxfordhb-9780199357505.indd 369
10/29/2013 11:10:23 PM
OUP UNCORRECTED PROOF – FIRSTPROOFS, Fri Oct 18 2013, NEWGEN
370
M ass Polit ics in Tough T imes
Vanhuysse, Pieter. 2006. Divide and Pacify: Strategic Social Policies and Political Protests
in Post-Communist Democracies. Budapest: Central European University Press.
Walton, John, and David Seddon, eds. 1994. Free Markets and Food Riots. The Politics of
Global Adjustment. London: Blackwell.
World Bank. 2006. Economic Policy and External Debt Data. http://data.worldbank.
org/indicator/NY.GDP.MKTP.KD.ZG.
oxfordhb-9780199357505.indd 370
10/29/2013 11:10:23 PM